On the Detection of Feature Points of 3D Facial Image and Its Application to 3D Facial Caricature

This paper proposes an automated method for extracting 39 feature point$ from 3D facial iniage, and shows that 3D facial caricaturing can be realized by such these smaller number of feature point. and that these 39 feature points could be extmcted automatically. As a result, it was clarified by comparing 3D caricature of 3D PICASSO system with that of a famous caricaturist Mr. Yoshida that 3D features of the facial image could be utilized to generate more impressive facial caricature then the 2D one. feature points, which are provided mainly by the hand-operated works. 'Ihis paper proposed an automatic method to detect feature points and 39 effective feature points could be extracted automatically. Consequently, it was clarified that 3D facial caricature could be realized, and that 3D facial caricature could be more impressive to the viewers than 2D one. In each chapter, in chap.2, the principle of PICASSO system wa$ summarized. In chap.3, the details of the method to extract feature points was presented, and in chap.4 3D facial caricaturing system was introduced and its validity was clarified experimentally. 2 "Mean face" assumption and PICASSO system